We have made a tremendous amount of progress understanding sensory function. But this understanding fails to account for responses to natural stimuli. We propose to develop a mechanistic understanding of how natural stimuli are encoded in the primate retina through a combination of directed experimentation and quantitative modeling. This will encompass three aims: (1) determining the circuit mechanisms that explain striking differences in responses of On parasol ganglion cells to natural and arti?cial stimuli; (2) determining the origin, properties, and functional signi?cance of the receptive ?eld surround of retinal ganglion cells; and, (3) developing and applying new tools to reveal the importance of cone adaptation in shaping ganglion cell responses to natural stimuli.